Safety systems and important components in nuclear power plants (NPP) are designed to give immense amount of security during accident scenarios. It is important to have the safety systems and important components in top conditions since they are subjected to various range of ageing mechanisms. To keep these systems and components operational and in reliable working condition, they are regularly inspected, repaired and replaced. The component that is impossible or economically unviable to replace is the reactor pressure vessel (RPV). The safety significance of the RPV is that it is the one component in NPP which cannot be allowed to fail under any circumstances, because the consequences would be impossible to mitigate effectively. Thus the integrity of RPV can potentially define the NPP life-limiting conditions. To increase the credibility of RPV integrity, there are higher margins in load bearing capacity that occur during the operation of NPP. However the RPV material properties are under slow degradation by neutron irradiation, thermal ageing and other ageing mechanisms. [1]
Neutron irradiation from reactor core causes very slowly material embrittlement within the RPV. Embrittled material has lost some of its ductility and experienced increase in ductile-to-brittle transition temperature. This temperature is also known as nil ductility temperature (NDT). When material temperature is below NDT it can fracture easily when deformed.
Under normal NPP conditions this will not be a significant risk since the coolant temperature in primary loop is always higher than the NDT of the RPV. [2]
In Loviisa NPP, accident scenarios have been mapped out where sudden cooling of the RPV can occur and the temperatures could fall below the NDT level. Typically these kind of thermal shock accident scenarios are caused by rapid cooling on the internal side of the RPV, but there are also rarer cases where the external side experiences the similar thermal shock. Even though these accident scenarios are very unlikely to happen, a deep understanding of these accidents are important for increasing the safety of the RPV.
6 1.2 Goals and delimitations
The purpose of this thesis is to study specifically the impact of external thermal shock on the Loviisa NPP RPV and investigate mitigation of thermal stresses by external thermal insulation. Thermal shocks are difficult phenomenon and acquiring a deep understanding of thermal shocks typically requires real experiments. Heat transfer experiments on external thermal shock have been researched in Lappeenranta University of Technology (LUT) by using heat transfer test facility that is located in the University. However in this study the real experiments were not performed and therefore the study of external insulation is limited to methods that can be calculated and simulated rather easily.
The goal is to identify the necessary requirements for thermal insulation materials so they can withstand the challenging conditions around the Loviisa RPV. The defined requirements can further be used in the future if more research and development is performed. Another goal of this thesis is to develop script capable of calculating temperature field distributions within RPV and the thermal insulation during rapid transient cooling. These temperature distribution results can be used as input data for RPV integrity calculations. The final decision in the regard of using the external thermal insulation is up to the owner of Loviisa NPP.
1.3 Structure of thesis
The chapter 2 introduces some background information about pressurized thermal shock (PTS), and briefly clarifies how and why PTS analyses are generally done in NPP. Chapter also includes more specific information about postulated external thermal shock in Loviisa NPP and the general ways of mitigation.
The third chapter is about thermal insulation that is proposed as one potential solution for mitigating the thermal shock. The requirements for potential insulation materials are introduced, along with some passed materials and rejected materials. The challenge in attaching the thermal insulation and other possible thermal insulation methods are also
7 included in this chapter.
The fourth chapter introduces temperature distribution analysis. The assumptions, boundary conditions, correlations, developed Matlab script and the validation of the script are found in this chapter. A brief review of current state of the heat transfer test facility located in LUT is included in the end of the chapter. The following chapter displays the results from the simulations of external thermal shock, including RPV with and without thermal insulation and steady state calculations for few selected cases. Finally the conclusions and further research suggestions are discussed.
8
2 PRESSURIZED THERMAL SHOCK
Material experiences a thermal shock when it is rapidly cooled down. The rapid temperature drop makes the material to contract due to thermal expansion and this causes thermal stresses within the material. In NPPs the event of rapid cooling in RPV is defined as pressurized thermal shock because the rapid cooling is often paired with higher internal pressure. PTS only by itself is not significant risk to the RPV; however the PTS under very specific conditions can challenge the RPV integrity.
The RPV is exposed to neutron radiations that are slowly resulting in localized embrittlement of RPV steel and weld materials. The risk is possible if RPV has defects of critical sizes due to embrittlement and if severe transients within NPP system were to occur it could have destructive impact by propagating defects rapidly through the vessel ultimately leading to fracture or even the destruction of RPV. Normally the defects of critical sizes are very unlikely to happen due to regular inspection of RPV integrity. The severe transients that can lead to RPV fracture are called PTS events. PTS events are divided into internal and external cooling of the RPV surface. PTS events also typically involve repressurization of the RPV. Combination of both under specific conditions gives a challenge to the integrity demands for RPV. [3, 4]
Probabilistic risk assessment (PRA) and engineering judgment has successfully identified various scenarios leading to PTS. Typical scenarios include loss of coolant accidents (LOCA), large secondary leaks, stuck open pressurizer safety or relief valve, primary to secondary leakage accidents, inadvertent actuation of high pressure injection or make-up systems and accident scenarios resulting in cooling of RPV from external side. When a scenario such as LOCA happens in primary system, the water level may drop rapidly. NPP operators and automatic systems provide reserve water in order to prevent overheating in the core. Provided water is generally much colder than the water in primary system. The combination of temperature drop and repressurization causes significant thermal stresses on the RPV wall that could potentially initiate propagation of existing defects. [1, 3]
Fortunately, the chances that defects of critical sizes exist in the embrittled RPV and weld material with the combination of severe PTS are very low probability event. The subtlety in nuclear safety is that even the most unlikely accident scenarios are assumed to happen
9
and that leads to the research and development which is making the nuclear energy more secure.
2.1 PTS Analysis procedure
The PTS analysis procedure is performed as series of sequential steps as shown in the flowchart in Figure 1. The procedure starts with selecting and defining the PTS sequence.
[1]
Figure 1. PTS analysis flowchart.
The selection of PTS transients (Figure 1, 1) is often based on identified accident scenarios in the safety analysis reports. The main goal is to identify accident scenarios that are direct PTS events themselves or are accidents with other consequences that can lead to PTS event. Different sequences in PTS analysis are frequently unit specific. Depending on the unit, all the relevant, meaningful and unique plant features are taken into consideration.
Typically some of the sequences are defined in terms of severity where PRA has been
10
used. Comprehensive probabilistic PTS studies are used to select the most important PTS sequences contributing to RPV failure risk. [1, 4]
Thermal hydraulic analyses (Figure 1, 2) are used to assist the transient selection process and to provide some necessary input data for analyzing RPV structural integrity. Analysis is typically done by specific thermal hydraulic code or combination of codes. The following parameters are provided by thermal hydraulic analyses:
Fluid temperature field in downcomer or external side of the RPV.
Primary circuit pressure.
Local heat transfer coefficients of wall-to-coolant.
Coolant temperature field and local heat transfer coefficients are replaced with inner surface temperatures of the RPV wall if the thermal hydraulic code is able to provide it. [1, 4]
Temperature and stress field calculations (Figure 1, 3) in the RPV wall during PTS transients are crucial for determining the integrity of the vessel. Calculations for stress fields are required for each time steps. Stresses are typically solved by numerical or analytical methods. Analytical methods are more used in specific justified cases. Solving stress fields by using the finite element method (FEM) is used in most cases. [4]
Fracture mechanics calculations (Figure 1, 4) are part of structural analysis. In structural analysis the aim is to evaluate stress intensity factors for postulated defects within the RPV that are under tension by thermal hydraulic transients. Fracture mechanics are based on static fracture toughness and they are used to estimate brittle failure in the RPV with postulated defects. In most defects and transient combinations the linear elastic fracture mechanics is sufficient approach, where the intensity factor Kl is acceptable. The values for intensity factor Kl are typically solved by numerical methods based on FEM where the postulated defects are included in the meshed geometry. [1, 4]
Integrity assessment (Figure 1, 5) is the final stage in the evaluation of PTS analysis. It includes evaluation of final results, safety factors and assessment of uncertainties in the results. [1, 4]
11 2.2 PTS in Loviisa NPP
PTS transient scenarios have been widely researched for Loviisa NPP units one and two.
Most of the relevant PTS sequences are internal and few are external. In Loviisa, the main factors that influenced the most in the selection of overcooling sequence for PTS analysis procedure were:
The probability of accident occurrence.
Occurrence of cold plumes in the downcomer
Repressurization of primary circuit.
The rate at which primary circuit is cooled down.
The final temperature of primary circuit.
Typical feature for PTS-analysis is that the conservative assumptions are usually the opposite when comparing to traditional safety analysis procedures in NPP. Considering the PTS in RPV, the situation is more severe when emergency core cooling system (ECCS) is working as planned and the injected ECCS water is as cold as possible. [5] The thermal hydraulic analyses in Loviisa have been completed by using APROS [6] simulation program supported with REMIX [7] thermal mixing simulations for cold plume scenarios.
2.2.1 Loviisa RPV
The purpose of RPV is to contain the reactor core, core shroud and the coolant. RPV is one of the most important components in NPP since it is practically irreplaceable and it has to withstand high temperature, pressure and neutron irradiation throughout NPP’s operational lifetime. It is also crucial for RPV to be able to withstand all relevant postulated accident scenarios.
LO1 and LO2 have almost identical RPV structure. Both RPVs were manufactured in Soviet Union and they consist of seven circular shells that are welded together. Both have the thickness of 140mm for RPV shell. The internal side of RPV has three-layered cladding for corrosion protection. The cladding has a thickness of 9-10mm. The exceptions in thicknesses are in the location of welds number 6 and 7, where the thickness is 205mm
12
without including cladding. Base materials for both RPVs are quenched and tempered low alloyed chrome-molybdenum-vanadium steel. [8, 9] Table 1 contains properties of base and weld materials of LO2 RPV in different temperatures. The rolled out overview of LO2 RPV and weld locations (right side in the figure) are presented in Figure 2.
Table 1. Base and weld material properties of LO2 RPV. [10]
Temperature [°C]
Parameter Unit 20 100 200 300
Thermal conductivity, k [W/mK] 40.2 39.8 38.8 37.9 Specific heat capacity, cp [J/kgK] 502
Density, ρ [kg/m3] 7800
Figure 2. Rolled out overview of LO2 RPV [8]
13 were also rearranged for achieving lower leakage for neutron irradiation to the RPV. Same actions with the exception of thermal annealing were taken in LO2 in order to mitigate embrittlement within the RPV. [11]
Due to neutron irradiation the NDT levels in RPV base material and weld material has increased. The NDT values for Loviisa RPV internal and external surfaces have been calculated and taken into account for the PTS studies. [12]
2.3 Postulated rapid cooling of the external side of RPV
One of the overcooling sequences in Loviisa NPP that is studied in Fortum is the rapid cooling on the external side of the RPV by the unexpected start-up of emergency spraying system (TQ-system). It was chosen to be used for thermal insulation studies of this work.
The sequence is a result from unexpected start-up of containment emergency spraying system while the NPP is operating at full power. The injected water by the TQ-system accumulates to the lower containment sump. Eventually the water reaches the bottom of reactor cavity through air conditioning channels. Water rises upwards in the 30 cm gap between RPV outer surface and concrete wall while rapidly cooling the side of RPV. The rapid cooling causes thermal shock to the external side of the RPV. If the cooling is strong enough to drop the temperature under NDT levels and there are existing defects within the RPV, the worst case is a fracture occurring through the RPV. [5]
The rising water and the following rapid cooling is causing biggest stress to the weld that is located in the beltline region making it more embrittled location than elsewhere on the external surface of the RPV. In order to increase the effect of thermal shock in the sequence and thus making the study more conservative, the water temperature is assumed to be as low as possible. This is achieved by assuming the accident to happen during winter
14
with lowest possible sea water temperature and also having pessimistic assumptions for process, e.g. assuming that only one TQ-pump is operating in both redundancies. Before the water by TQ-system is sprayed to the containment it is cooled by intermediate circuit with seawater. Fewer operational TQ-pumps will lead to lower mass flow rate and it will further decrease the temperature of sprayed water. In addition two defects are assumed, where one is located in the internal side beneath the cladding and second one in the external side within the weld. The accident sequence is assumed to last for thirty minutes and then the TQ-system is halted by the operator. [5] The gap between RPV and concrete wall where the water accumulates is illustrated in Figure 3.
Figure 3. Lower part of the RPV. The gap between RPV outer surface and the concrete wall is shown by blue arrows. [8]
15
2.3.1 Conditions and transient progression
The TQ-system is assumed to start accidently when the NPP is operating at full power. The sequence lasts for 1800 seconds (30 minutes) and then TQ-system is assumed to be terminated by the operator. The starting conditions that are notable are the following:
Reactor power 100% (1500 MW)
Primary circuit pressure in pressurizer about 12.4 MPa
Water level in pressurizer 4.6 m
Temperature in hot leg 299.9 °C
Temperature in cold leg 266.1 °C
Coolant mass flow in primary circuit about 8500 kg/s
Temperatures in TQ-lines 16 °C
Total mass flow in TQ-system 400 kg/s
Temperature of seawater 0.1°C
Accident sequence has been analyzed by Fortum with APROS process simulation software. [13] Following figures are some of the results from the simulation. The estimation for water elevation in the gap between RPV outer surface and concrete wall is shown in Figure 4.
16
Figure 4. Water level between RPV outer surface and concrete wall in postulated cooling sequence. [13]
The estimation of surface temperature on the weld that is located in the beltline region is shown in Figure 5 and the corresponding heat transfer coefficient from wall to water during the transient is found in Figure 6.
17
Figure 5. External surface temperature at the weld located in the beltline region. [13]
Figure 6. Heat transfer coefficient from RPV surface to water at the weld located in the beltline region. [13]
18 2.4 Reduction of PTS
Some methods executed in Loviisa NPP in order to reduce PTS were introduced in section 2.2.1. Mitigation of radiation embrittlement leading to slower rising of NDT in the future is an effective preemptive method. The actions taken for the reduction of PTS need to be properly adjusted on specific NPP case by case. This section introduces methods for mitigating PTS.
2.4.1 General methods
General mitigation methods can be roughly divided into two categories. First category includes the methods that are directly or indirectly resulting to better RPV integrity and lower NDT levels. Second category includes methods that are directly reducing or eliminating the risks or probabilities of PTS.
First category that improves the RPV integrity is mostly achieved by reducing the impact of radiation embrittlement. Actions taken in Loviisa that fall in the first category and recommended methods by IAEA include [14, 2]
Optimizing fuel management by using low leakage core loading pattern in order to reduce neutron flux to the RPV wall.
Reconfiguration of the core by using partial shielding assemblies or dummy elements such as hafnium or stainless steel to reduce neutron flux at wanted areas of the RPV wall.
Thermal annealing of the embrittled RPV for the purpose of recovering material integrity.
Overall benefits of neutron flux reduction are depending on time of implementation, original neutron flux levels and chemical composition of RPV material. All mitigation actions should be implemented and researched properly case by case on each considered NPP. [1] Second category includes actions that directly have mitigation impact on PTS:
Reducing the effect of thermal shock by raising water temperature in ECCS.
Removing the threat of cold plumes by adjusting the injection of ECCS to primary coolant in a manner that completely mixed flowing conditions are achieved.
Shut-off head and injection capacity adjustment in high pressure injection pumps.
19
Adjustment of steamline isolation procedures.
Operator training and improvement of the emergency situation instruction manuals regarding PTS risks.
2.4.2 Thermal insulation of sensitive weld
A proposal for more specialized solution for mitigating thermal shock is by thermally insulating most sensitive parts of the RPV. Typically the welds around the beltline region suffer the most from embrittlement. Thermal insulation on internal side of the RPV might be utmost challenging because of more difficult conditions e.g. higher radiation, corrosion, attachment and installation challenges and insulation effect on the existing cladding.
However externally the thermal insulation encounters fewer challenges and could therefore be one potential option for reducing the impact of thermal shocks
In the external overcooling cases such as mentioned in section 2.3 the thermal insulation could reduce the thermal shock effectively by mitigating the temperature gradient in the RPV during accident scenarios. Mitigation of temperature gradient in overcooling accident
In the external overcooling cases such as mentioned in section 2.3 the thermal insulation could reduce the thermal shock effectively by mitigating the temperature gradient in the RPV during accident scenarios. Mitigation of temperature gradient in overcooling accident